Unravelling the elicitin perception pathway against Phytophthora in potato

Potato (Solanum tuberosum) is the most important non-cereal crop for human consumption and its starch and fibers are also used to produce several industrial products. Potato suffers from many pests and diseases. The most threatening and re-emerging disease of potato worldwide is late blight, which is caused by the notorious oomycete Phytophthora infestans (Chapter 1). This pathogen is a hemi-biotrophic organism that secretes a huge arsenal of apoplastic and host-translocated cytoplasmic effectors in order to colonize the host. Interestingly, wild potato plants have evolved receptors that recognize some of those effectors and trigger defense responses. These wild plants are a major source of resistance genes that can be transferred to the cultivated potato. Resistant cultivars obtained by breeding are highly desirable because the costs for chemical control of the disease are very high and there is also a need to reduce their use in order to preserve the environment. Moreover, the pathogen has developed resistance to some of those chemicals.

Receptors that recognize cytoplasmic effectors often belong to the nucleotide-binding leucine-rich repeat (NLR) family of resistance genes, (R genes), and they have been used for several years in resistance breeding. Despite the effectivity of NLRs in providing resistance, so far, most single NLR genes introduced into cultivated potato have been defeated by P. infestans rather quickly. To effectively control the pathogen in the long term, resistance gene stacking approaches in combination with new layers of defense have to be considered.

Like NLRs, cell surface-residing receptors, or pattern recognition receptors (PRRs) trigger defense responses and contribute to basal or non-host resistance against pathogens. They do so by recognizing apoplastic effectors or microbe-associated molecular patterns (MAMPs). Usually PRR-triggered resistance is quantitative and not as robust as NLR-based resistance, however it is believed to be more durable. In potato, PRR-based immunity that is triggered by the recognition of MAMPs/effectors has remained unexplored. The first receptor against oomycetes was recently identified from a wild potato species, S. microdontum. The receptor, named ELICITIN RESPONSE (ELR), was found to recognize elicitins, a family of conserved apoplastic effectors with MAMP features, found specifically in Phytophthora and Pythium (Chapter 1). The research described in this thesis focused on studying PRR-triggered immunity in potato against the devastating pathogen P. infestans, by using ELR as a model (Chapter 1).

Effector-assisted breeding has proven to be a great tool for identifying resistances against pathogens and was pioneered from research on potato late blight. In a process also known as effector genomics (effectoromics), candidate effectors are predicted from the genome of pathogens such as P. infestans. Candidate effectors are cloned in plant expression vectors and are screened in a wild resistant germplasm for occurrence of specific responses. In Chapter 2 we show how effectoromics can be used to identify PRRs in Solanum spp. Simplified protocols are described for performing the effector screens, selecting plants for crosses and genetically mapping the responses. For performing effector screens, besides the routinely used potato virus X (PVX) agroinfection and agroinfiltration, we describe the use of recombinant apoplastic effector proteins. This strategy can complement the results obtained by the other routinely applied Agrobacterium-based methods or enable screening of Agrobacterium- or PVX-recalcitrant plants. We provide protocols for heterologous apoplastic effector expression in the yeast Pichia pastoris. This includes recombinant effector design, cloning, high throughput P. pastoris clone selection and small scale protein production. We also provide an example with the production of six different P. infestans effectors using this system.

ELR is a receptor-like protein (RLP) and as such, it lacks a cytoplasmic signaling domain that is required for triggering defense responses. It was known that SOMATIC EMBRYOGENESIS RECEPTOR KINASE 3 (SERK3) is a receptor-like kinase (RLK) required for cell death triggered by the elicitin INF1 and that it also biochemically associates with ELR. In Chapter 3 we investigated the association of ELR with another interactor, which has been proposed to be specific for RLPs; the RLK SUPPRESSOR OF BIR1-1 (SOBIR1). Virus-induced gene silencing (VIGS) assays showed that SOBIR1 is required for cell death response triggered by INF1 and for basal resistance to P. infestans in Nicotiana benthamiana. Genetic complementation assays demonstrated that the kinase of SOBIR1 is required for INF1-triggered cell death. Protein co-immunoprecipitation studies showed that ELR is associating with S. microdontum SOBIR1 and its close homolog SOBIR1-like. From our findings it seems that ELR is found in a constitutive complex with SOBIR1, which recruits SERK3 upon INF1 elicitation.

True PRRs are known to physically interact with their ligands, however, this was not explored yet since the identification of ELR. In Chapter 4 we studied whether ELR is able to interact with various elicitins by using in planta and in vitro co-immunoprecipitation assays. We showed that ELR is able to physically bind INF1 in both setups, indicating that the interaction is very specific. Moreover, we found that ELR binds with the elicitins ParA1 (P. parasitica) and β-CRY (P. cryptogea). We also found that ELR is able to trigger cell death with additional elicitins, but likely binds them with lower affinity that remained under the detection limit. Interestingly, we observed that one C-terminally tagged version of INF1 was binding to ELR but failed to trigger cell death, in contrast to N-terminally tagged INF1. Similarly, a β-CRY dimer failed to trigger cell death when infiltrated in leaves expressing ELR. We hypothesized that the cause of these phenotypes could be due to altered interaction of ELR with the RLKs SOBIR1 or SERK3. Indeed, with the C-terminally tagged INF1, we found that both SOBIR1 and SERK3 were not in complex with ELR, while the β-CRY dimer was not allowing SOBIR1 to associate. To our knowledge, this is the first time that such a co-receptor inhibition is reported. Our data, therefore, highlight the necessity of performing functional control experiments for tagged proteins and further strengthen our earlier finding that both SOBIR1 and SERK3 are required for INF1-triggered cell death by ELR. These discoveries could prove useful for enabling detailed studies on RLP signaling.

When transformed into cultivated potato, ELR was known to enhance resistance against P. infestans, however, it remained unknown what the resistance contribution of this PRR in its native genetic background (i.e. S. microdontum) is. We hypothesized that recent developments of genome editing technologies can be used to perform such studies in non-model plants like (wild) potato. In Chapter 5, we used clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9)-mediated genome editing to target ELR in wild Solanum. We generated a CRISPR/Cas9 construct that was found to be effective in inducing targeted mutations in ELR when tested in a transient setup in planta. Since it was impossible to transform the wild S. microdontum with the construct, we attempted the transformation of five different species that were known to carry ELR homologs. We obtained transformants from S. edinense, S. papita, S. phureja and S. chacoense, which were then screened for altered responses to INF1. Promising transformants with altered responses to INF1 were obtained in the pentaploid S. edinense and tetraploid S. papita. These transformants were subsequently searched for mutations using a mutation enrichment approach, followed by PCR and sequencing. No full ELR knock-outs were obtained, however, partly mutated and partly wildtype alleles could be found in these transformants. Altogether, these findings show that CRISPR/Cas9 approaches are effective for functional characterization of genes in wild Solanum spp. However, they are limited by the transformation efficiency and ploidy level of a said genotype.

Finally, in Chapter 6, the findings of this thesis are discussed and placed in a broader perspective. A schematic overview of the perception of elicitins by ELR in wild potato is provided including all of our findings on elicitin binding, interaction with co-receptors SERK3 and SOBIR1, as well as some preliminary findings on AVR3a, an effector that suppresses INF1-triggered cell death as well as receptor endocytosis. In addition we propose two proteomic approaches for PRR identification, taking advantage of the obtained knowledge on effector/PRR or PRR/co-receptor interactions. Overall my research has contributed to the characterization of the first line of induced defense against Phytophthora in potato and could be instrumental for achieving durable resistance against late blight.